Underfloor radiant - air gap or no air gap?

Paul E.

I recently was checking out RTI's Radiant Trak aluminum plate installation instructions and they say to install insulation in direct contact with the plates, contrary to what I've seen generally that says to allow an air gap of a couple inches for better performance.

In my case I plan on installing 1/2" PEX on 8" centers using RTI's extruded aluminum plates. I would love to be able to get away with a single layer of the low profile "bubble pack" type foil insulation in direct contact with the plates. This would give me a nice space margin from the romex runs passing thru mid-joist. Is that type of foil insulation enough, or should it be backed by another layer of R-11/R-13 fiberglass?

Does anybody know why RTI seems to differ from others on the air gap issue?

Thanks.

cruizer

I think it may be 6 of one, half dozen of the other. If you have an airgap between the plates and insulation (about 2") Your system reaction time will be slightly slower, because it is going to fill the air gap before trying to move through the subfloor. But you will get a more even heating throughout the joist cavity. If you put the insulation up against the plates, your will have a quicker reaction time, as the heat will be directed straight up through the subfloor. But you run the change of having less heat (like a very small band of striping) between the plates, more then likely that won't even be noticable.

As for foil insulation. I believe that the reflection aspect of it would work up until the area between it and the heat source was saked up with heat. Once the airspace was soaked with heat, I think that the heat will transfer right through the foil due to conduction, because technically it's not reflecting anymore.

I believe it would work if you put addition foam or batt insulation behind it.

Eric Taylor_5

My Take

Given three types of heat transfer, if you remove one or more it HAS to reduce the transfer. So, if you leave an air gap between the insulation and the plates, then there can't be any heat conducted to the insulation directly. Not that it is a good conductor to start with, but to be thorough you have to consider it. You are left with convection and radiation. We know that warm air rises, so convection losses downward in a dead air space should be small. Block the radiation with foil and you should be good to go. Think of a thermos bottle. Foil lined for radiation, made of glass instead of metal to reduce conduction, and a vacuum to virtually eliminate convection and conduction through the wall.

I can't imagine any possible way for an air gap to increase the losses.

Eric

Tim Doran

Air Space

The 2" air space is only need when using pex clips, hangers or other suspended methods. When using plates the insulation should be up against the plate. The idea with plates is to rely on conduction and to keep from heating the air space in the joist bay. Hope this helps.

Tim D

Bill Wright_2

Tim's right, RTI is not alone ...

Many manufacturers do not require an air gap for staple up w/ plates.

If you do not have plates, then the air gap (with the foil backed insulation) improves your output. If you have plates, then the air gap actually lowers your output (you're using energy to heat air, instead of to heat the subfloor).

I've attached the JLC article summarizing Dr. Chapman's (KSU) ASHRAE study on this very question.

Bill W.

Dave Holdorf

Air Gap and water temp

The idea of the insulation up tight is to get our radiant systems to operate at the lowest possible required water temp. Therefor the "striping" can be almost eliminated and the response time can increased.

Thanks

Dave H.

Newt Loken

Plates/air-space

It seems to me it may vary depending on the style of plate. The extruded plates are stiffer anyway and probably have a higher contact area than "sheet/pressed" (?) plates like Wirsbos. Stapling on one side of a Wirsbo plate leaves a significant area hanging, sorta like a baseboard aluminum fin wing (spelled "convection"). Besides, the heat capacity of air is...next to nothing when compared to wood sub, and finished floor with a bit of moisture in it. I vote for atleast some air space to create a more homogeneous temperature. Of course when the joist dimension is skinny push it on up....
Newt